Chapter 3 – Research
Research: a systematic way of finding answers to questions. It’s a method of inquiry that follows
certain rules.
A Scientific Approach – Section Summary
- A scientific approach to abnormal child psychology is a way of thinking about how best to
understand and answer questions of interest, not just an accumulation of specific methods,
practices, or procedures.
- Science requires that theories be backed up by evidence from controlled studies and that
observations be checked and repeated before conclusions are drawn.
- Facilitated communication (FC) meets many of the criteria of pseudoscience because
demonstrations of benefit are based on anecdotes or testimonials, the child’s baseline abilities
and the possibility of spontaneous improvement are ignored, and typical scientific procedures are
disavowed.
- What distinguishes science from pseudoscience is that scientists play by the rules of science, and
prepared to admit when they are wrong, and are open to change.
The research process
Research in abnormal child psychology is beset characterized as a multistage process involving
key decisions at various points.
• Step 1: researcher(s) developing a hypothesis on the basis of observation, theory and
previous findings, and deciding on a general approach to research.
• Step 2: identifying the sample to be studies, selecting measurement methods, and
developing a research design and procedures.
o The research design and procedures must balance practical considerations with the
adequacy of the research to address the hypothesis under investigation.
• Step 3: gathering and analyzing the data and interpreting the results in relation to theory
and previous findings in an attempt to resolve the problem in step 1.
Epidemiological research: the study of incidence, prevalence, and co-occurrence of childhood
disorders and competencies in clinic-referred and community samples.
Incidence rates: the rate at which new cases of a disorder appear over a specified period of time.
Prevalence rates: the number of cases of a disorder, whether new or previously existing, that are
observed during a specified period of time.
Knowledge about the risk for, and expression of, an individual disorder over the life course helps
us understand the nature of the disorder and use this understanding as the basis for prevention
and treatment.
Estimates of prevalence vary widely:
• Depending on whether children from clinics or children from community samples are
the focus of the study.
• Depending on whether cases are defined in terms of patterns of symptoms, impairment
in functioning, or both.
• It varies in relation to demographic and situational factors, such as socioeconomic status
(SES), parents’ marital status, the child’s age, gender, and cultural background, etc.
Correlates: refers to variables that are associated at a particular point in time with no clear proof
that one precedes the other.
Causes: they influence, directly or indirectly, through other variables, the occurrence of a
behavior or disorder of interest.
, Research into risk and protective factors often requires that large samples of children be studies
and that multiple domains of child functioning (physical, cognitive, and psychosocial) be
assessed over long periods of time. This is necessary because:
• Only a small proportion of children at risk for a problem will actually develop the
disorder (therefore a large sample of children is required).
• The areas of child functioning that will be affected, and how they will be affected, are not
known in advance.
• The ages at which a disorder may occur or reoccur are also not known in advance.
Moderator variables: a factor that influences the direction or strength of a relationship between
variables.
Mediator variables: the process, mechanism, or means through which a variable produces a
specific outcome.
à Key difference between these two variables: moderators have an independent effect on the
existing relationship between two variables, whereas mediators account for some or all of the
apparent relationship between two variables.
Randomized controlled trials (RCTs): a design used to evaluate treatment outcomes in which
children with a particular problem are randomly assigned to various treatment and control
conditions.
Treatment efficacy: the degree to which a treatment can produce changes under well-controlled
conditions that depart from those typically used in clinical practice.
Treatment effectiveness: the degree to which a treatment can be shown to work in actual clinical
practice, as opposed to under controlled laboratory conditions.
The Research Process – Section Summary
- Research is a multistage process that involves generating hypotheses, devising an overall plan,
selecting measures, developing a research design and procedures, gathering and analyzing the
data, and interpreting the results.
- One’s theory of abnormal child behavior determines the variables studies, the choice of research
methods, and the interpretation of research findings.
- Questions about the nature and distribution of childhood problems are addressed through
epidemiological research into the incidence and prevalence of childhood disorders and
competencies in clinic-referred and community samples.
- Other common research topics in abnormal child psychology focus on correlates, risk and
protective factors, causes, moderating and mediating variables, outcomes, and interventions for
childhood disorders.
Methods of studying behavior – Standardization, reliability, and validity
Standardization: the process by which a set of standards or norms is specified for a
measurement procedure so that it can be used consistently across different assessments.
• Used to ensure consistency in the way it is used by different assessors and across measurements
occasions.
Reliability: the extent to which the result of an experiment is consistent or repeatable.
• Internal consistency: whether all parts of a method of measurement contribute in a
meaningful way to the information obtained.
• Interrater reliability: information must not depend on a single observer or clinician;
various people must agree on what they see.
• Test-retest reliability: tests or interviews repeated within a short time interval should
yield similar results on the two occasions. Thus, the results need to be stable over time.
Research: a systematic way of finding answers to questions. It’s a method of inquiry that follows
certain rules.
A Scientific Approach – Section Summary
- A scientific approach to abnormal child psychology is a way of thinking about how best to
understand and answer questions of interest, not just an accumulation of specific methods,
practices, or procedures.
- Science requires that theories be backed up by evidence from controlled studies and that
observations be checked and repeated before conclusions are drawn.
- Facilitated communication (FC) meets many of the criteria of pseudoscience because
demonstrations of benefit are based on anecdotes or testimonials, the child’s baseline abilities
and the possibility of spontaneous improvement are ignored, and typical scientific procedures are
disavowed.
- What distinguishes science from pseudoscience is that scientists play by the rules of science, and
prepared to admit when they are wrong, and are open to change.
The research process
Research in abnormal child psychology is beset characterized as a multistage process involving
key decisions at various points.
• Step 1: researcher(s) developing a hypothesis on the basis of observation, theory and
previous findings, and deciding on a general approach to research.
• Step 2: identifying the sample to be studies, selecting measurement methods, and
developing a research design and procedures.
o The research design and procedures must balance practical considerations with the
adequacy of the research to address the hypothesis under investigation.
• Step 3: gathering and analyzing the data and interpreting the results in relation to theory
and previous findings in an attempt to resolve the problem in step 1.
Epidemiological research: the study of incidence, prevalence, and co-occurrence of childhood
disorders and competencies in clinic-referred and community samples.
Incidence rates: the rate at which new cases of a disorder appear over a specified period of time.
Prevalence rates: the number of cases of a disorder, whether new or previously existing, that are
observed during a specified period of time.
Knowledge about the risk for, and expression of, an individual disorder over the life course helps
us understand the nature of the disorder and use this understanding as the basis for prevention
and treatment.
Estimates of prevalence vary widely:
• Depending on whether children from clinics or children from community samples are
the focus of the study.
• Depending on whether cases are defined in terms of patterns of symptoms, impairment
in functioning, or both.
• It varies in relation to demographic and situational factors, such as socioeconomic status
(SES), parents’ marital status, the child’s age, gender, and cultural background, etc.
Correlates: refers to variables that are associated at a particular point in time with no clear proof
that one precedes the other.
Causes: they influence, directly or indirectly, through other variables, the occurrence of a
behavior or disorder of interest.
, Research into risk and protective factors often requires that large samples of children be studies
and that multiple domains of child functioning (physical, cognitive, and psychosocial) be
assessed over long periods of time. This is necessary because:
• Only a small proportion of children at risk for a problem will actually develop the
disorder (therefore a large sample of children is required).
• The areas of child functioning that will be affected, and how they will be affected, are not
known in advance.
• The ages at which a disorder may occur or reoccur are also not known in advance.
Moderator variables: a factor that influences the direction or strength of a relationship between
variables.
Mediator variables: the process, mechanism, or means through which a variable produces a
specific outcome.
à Key difference between these two variables: moderators have an independent effect on the
existing relationship between two variables, whereas mediators account for some or all of the
apparent relationship between two variables.
Randomized controlled trials (RCTs): a design used to evaluate treatment outcomes in which
children with a particular problem are randomly assigned to various treatment and control
conditions.
Treatment efficacy: the degree to which a treatment can produce changes under well-controlled
conditions that depart from those typically used in clinical practice.
Treatment effectiveness: the degree to which a treatment can be shown to work in actual clinical
practice, as opposed to under controlled laboratory conditions.
The Research Process – Section Summary
- Research is a multistage process that involves generating hypotheses, devising an overall plan,
selecting measures, developing a research design and procedures, gathering and analyzing the
data, and interpreting the results.
- One’s theory of abnormal child behavior determines the variables studies, the choice of research
methods, and the interpretation of research findings.
- Questions about the nature and distribution of childhood problems are addressed through
epidemiological research into the incidence and prevalence of childhood disorders and
competencies in clinic-referred and community samples.
- Other common research topics in abnormal child psychology focus on correlates, risk and
protective factors, causes, moderating and mediating variables, outcomes, and interventions for
childhood disorders.
Methods of studying behavior – Standardization, reliability, and validity
Standardization: the process by which a set of standards or norms is specified for a
measurement procedure so that it can be used consistently across different assessments.
• Used to ensure consistency in the way it is used by different assessors and across measurements
occasions.
Reliability: the extent to which the result of an experiment is consistent or repeatable.
• Internal consistency: whether all parts of a method of measurement contribute in a
meaningful way to the information obtained.
• Interrater reliability: information must not depend on a single observer or clinician;
various people must agree on what they see.
• Test-retest reliability: tests or interviews repeated within a short time interval should
yield similar results on the two occasions. Thus, the results need to be stable over time.
